Resilience Modeling in Distributed Systems Based on the Generalized Erdős – Rényi Model and the Gilbert – Elliott Model
( Pp. 79-88)

The aim of this study is to develop and validate a model for assessing the resilience of distributed systems that considers both the structural characteristics of the network and the dynamic behavior of connections. The proposed model combines graph analysis (based on the Erdős – Rényi model) and statistical modeling (Gilbert – Elliott model), integrating connectivity probabilities and successful connection metrics to evaluate network resilience. The primary objective was to create an approach capable of accurately describing real-world network processes and identifying potential points of degradation. The model was tested using a local Kubernetes cluster, where a test CRUD service was deployed under load for 24 hours. Collected metrics, including packet loss, latency, and throughput, were compared to the model’s predictions. The results showed minimal deviations between theoretical predictions and empirical data, confirming the model’s adequacy. The study concludes that the proposed approach can not only accurately describe current network processes but also serve as a foundation for decision-making regarding scaling and replication. The model’s flexibility ensures its relevance in scenarios involving changes in network topology or connection quality, making it applicable for analyzing modern distributed systems.

Sukhoplyuev D.I., Nazarov A.N. Resilience Modeling in Distributed Systems Based on the Generalized Erdős – Rényi Model and the Gilbert – Elliott Model. Computational Nanotechnology. 2025. Vol. 12. No. 1. Pp. 79–88. (In Rus.). DOI: 10.33693/2313-223X-2025-12-1-79-88. EDN: MKDXXJ

Evstratov V.V. Container orchestration using Kubernetes. Young Scientist. 2020. No. 51 (341). Pp. 11–13. (In Rus.). EDN: DEALJL.
Gu L., Xu H., Li Z. et al. Container session level traffic prediction from network interface usage. IEEE Transactions on Sustainable Computing. 2023. Vol. 8. No. 3. Pp. 400–411. DOI: 10.1109/tsusc.2023.3252595. EDN: OYVTND.
Kovalev M.G. Tracing network packets in the Linux Kernel using eBPF. Proceedings of the Institute for System Programming of the RAS. 2020. Vol. 32. No. 3. Pp. 71–78. DOI: 10.15514/ISPRAS-2020-32(3)-6. EDN: AKNDPV.
Ibryam B. Evolution of cloud computing: From raw infrastructure to serverless applications. Internet from the Inside. 2023. No. 19. Pp. 9–13. (In Rus.) EDN: VQDCGX.
Fokin A.B. Method for calculating connectivity probabilities (availability coefficients) in a telecommunications network supporting fault tolerance mechanisms. Information Systems and Technologies. 2023. No. 4 (138). Pp. 83–91. (In Rus.). EDN: CWQJBV.
Hablinger G., Hohlfeld O. The Gilbert – Elliott model for packet loss in real time services on the Internet. In: 14th GI/ITG Conference on Measuring, Modelling and Evaluation of Computer and Communication Systems, MMB 2008 (Dortmund, 2008). Dortmund, 2008. P. 5755057. EDN: SSWAZB.
Gostev I.M., Golosov P.E. Efficiency analysis of a cloud computing system serving a flow of tasks with deadline constraints under multiple server failures. Software Engineering. 2023. Vol. 14. No. 6. Pp. 278–284. (In Rus.). DOI: 10.17587/prin.14.278-284. EDN: EGXWYI.
Ivutin A.N., Novikov A.S., Pestin M.S., Voloshko A.G. Decentralized protocol for organizing sustainable interaction of subscribers in networks with high topology dynamics. Informatics and Automation. 2024. Vol. 23. No. 3. Pp. 727–765. (In Rus.). DOI: 10.15622/ia.23.3.4. EDN: JSQFAC.
Oblakova T.V., Kasupovich E. Numerical study of persistent time series based on the ARFIMA model. Mathematical Modeling and Numerical Methods. 2022. No. 4 (36). Pp. 114–125. (In Rus.). DOI: 10.18698/2309-3684-2022-4-114125. EDN: MTJIKO.
Pattanayak R.M.P., Sangameswar M.V., Vodnala D., Das H. Fuzzy time series forecasting approach using LSTM model. Computacion y Sistemas. 2022. Vol. 26. No. 1. DOI: 10.13053/cys-26-1-4192. EDN: JNBLPQ.
Zayats O.I., Korenevskaya M.M., Ilyashenko A.S., Mulyukha V.A. Mass service system with absolute priority, probabilistic push-out mechanism, and repeat requests. Informatics and Automation. 2024. Vol. 23. No. 2. Pp. 325–351. (In Rus.). DOI: 10.15622/ia.23.2.1. EDN: KXQKLM.
Rumyantsev A.S., Dolgaleva D.S., Golovin A.S. Study of stationary characteristics of multiserver models with redundancy. Software Systems: Theory and Applications. 2023. Vol. 14. No. 1 (56). Pp. 55–94. (In Rus.). DOI: 10.25209/2079-3316-2023-14-1-55-94. EDN: CMFWJG.
Melnikov B.F., Terentyeva Yu.Yu. Practical applications of the Floyd – Warshall algorithm and its modifications. Informatization and Communication. 2023. No. 5. Pp. 7–14. (In Rus.). DOI: 10.34219/2078-8320-2023-14-5-7-14. EDN: YPIUSI.
Asratyan R.E. Deterministic model of packet request processing in a multithreaded server. Software Engineering. 2023. Vol. 14. No. 4. Pp. 155–164. (In Rus.). DOI: 10.17587/prin.14.155-164. EDN: ZCWJCH.
Allakin V.V., Budko N.P., Vasiliev N.V. General approach to building prospective monitoring systems for distributed information and telecommunication networks. Management, Communication, and Security Systems. 2021. No. 4. Pp. 125–227. (In Rus.). DOI: 10.24412/2410-9916-2021-4-125-227. EDN: JPFJRO.

distributed systems, graph theory, Gilbert – Elliott model, network monitoring.